The objective of this project is to discover proteins in human serum that can be used as biomarkers for the early detection of ovarian cancer. Ultimately, the goal would be to utilize these proteins to develop a simple blood test that is capable of diagnosing ovarian cancer at its earliest stages, when there is the greatest likelihood of succesful treatment. In order to find the proteins in cancer serum that could be indicative of a diseased state, selection of recombinant single chain variable fragment antibodies (scFv) from a library of over 100 million unique scFv will be performed. For this, several rounds of selection for proteins present in cancer serum and against proteins found in normal serum will be done. After selection, an array with up to 10000 unique scFv spots will be printed. Each microarray will be challenged with a pair of matched serum samples, one from ovarian cancer (proteins labeled with a red dye) and one from a normal control (labeled with a green dye). The differences in the amounts of each of the antigens that are bound to the scFv will be quantified as a fluorescence ratio. Antigens that show differential expression in cancer vs. normal samples will be studied further for their biological function in ovarian cancer for their validation as biomarkers.